[1]魏臻珠,刘明宇,王昱龙,等.基于风场空间与气象融合的风电集群短期功率预测[J].郑州大学学报(工学版),2026,47(5):26-34.[doi:10.13705/j.issn.1671-6833.2026.02.005]
　WEI Zhenzhu,LIU Mingyu,WANG Yulong,et al.Short-term Power Prediction of Wind Power Clusters Based on Wind Field Spatial and Meteorological Fusion[J].Journal of Zhengzhou University (Engineering Science),2026,47(5):26-34.[doi:10.13705/j.issn.1671-6833.2026.02.005]
点击复制

基于风场空间与气象融合的风电集群短期功率预测()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
47
期数:
2026年5期
页码:
26-34
栏目:
出版日期:
2026-09-09

文章信息/Info

Title:
Short-term Power Prediction of Wind Power Clusters Based on Wind Field Spatial and Meteorological Fusion
文章编号:
1671-6833(2026)05-0026-09
作者:
魏臻珠1, 刘明宇1, 王昱龙1, 周 燕2, 蒋建东1
1. 郑州大学 电气与信息工程学院,河南 郑州 450001;2. 国网河南省电力公司洛阳供电公司,河南 洛阳 471000
Author(s):
WEI Zhenzhu1, LIU Mingyu1, WANG Yulong1, ZHOU Yan2, JIANG Jiandong1
1. School of Electrical and Information Engineering, Zhengzhou University, Zhengzhou 450001, China; 2. Luoyang Power Supply Company of State Grid Henan Electric Power Company, Luoyang 471000, China
关键词:
风电集群功率预测;  时空图卷积神经网络;  多头自注意力机制;  图数据结构;  深度学习
Keywords:
wind power cluster forecasting;  spatio-temporal graph convolutional neural network;  multi-head self-attention mechanism;  graph data structure;  deep learning
分类号:
TM614TP183
DOI:
10.13705/j.issn.1671-6833.2026.02.005
文献标志码:
A
摘要:
针对传统风电集群功率预测方法在空间上未有效考虑场站间气象关联特性，难以基于单场预测高效推演集群整体功率的问题，且为了充分挖掘离散型数值气象预报（NWP）异构气象信息的时空耦合复杂特性嵌入表征，提出一种基于注意力时空嵌入机制的多维度场站时空信息融合框架。首先，利用多头自注意力机制（multi‑head self‑attention）直接融合空间特征，增强模型对多场站空间功率关联的捕捉能力。其次，基于最大信息系数解构集群位置信息，构建反映气象关联的非欧几里得图数据结构，并结合时空注意力机制实现场站及其邻域时空特征的交叉融合，动态调整场站间影响权重以捕捉时空动态相关性。并通过编‑解码器架构将空间与时间特征集成至统一语义空间，捕获序列时间连续性。最后，基于中国西北地区实际风电场运行数据对所提模型进行验证。实验结果表明：所提方法RMSE与MAE两项误差评价指标比其他6种预测模型均显著降低，有效验证其先进性与适应性。
Abstract:
Given that traditional wind power cluster prediction methods failed to effectively account for the spatial meteorological correlations among stations and struggle to efficiently deduce the overall cluster power based on single‑station predictions, in this study a multi‑dimensional spatiotemporal information fusion framework was proposed for stations based on an attention‑based spatiotemporal embedding mechanism. This framework aimed to fully exploit the complex spatiotemporally coupled characteristics embedded within discrete numerical weather prediction (NWP) heterogeneous meteorological information. Firstly, a multi‑head self‑attention mechanism was employed to directly fuse spatial features, enhancing the model’s ability to capture spatial power correlations across multiple stations. Secondly, cluster location information was deconstructed using the maximal information coefficient to construct a non‑Euclidean graph data structure reflecting meteorological correlations. This was combined with a spatial‑temporal attention mechanism to achieve cross‑fusion of spatiotemporal features between stations and their neighborhoods, dynamically adjusting the influence weights among stations to capture spatiotemporal dynamic dependencies. Furthermore, an encoder‑decoder architecture was used to integrate spatial and spatiotemporal features into a unified semantic space to capture temporal continuity within sequences. Finally, the proposed model was verified based on the actual wind farm operation data of a certain region in Northwest China. Experimental results showed that the two error evaluation indexes of the proposed method RMSE and MAE were significantly lower than those of the other six prediction models, which effectively verified its advancement and adaptability.

参考文献/References:

[1] Shao Guiping, Xu Honghua. Research on present situation and future development trend of renewable energy‑integrated system[J]. Solar Energy, 2024(7): 127‑132.［邵桂萍，许洪华. 可再生能源综合系统现状与未来发展趋势研究［J］. 太阳能，2024（7）：127‑132.］
[2] Solé J, García‑Olivares A, Turiel A, et al. Renewable transitions and the net energy from oil liquids: a scenarios study[J]. Renewable Energy, 2018, 116: 258‑271.
[3] Zhang Wenyu, Ma Keke, Guo Zhenhai, et al. Multistep prediction of wind speed based on grey wolf algorithm and extreme learning machine[J]. Journal of Zhengzhou University (Engineering Science), 2024, 45(2): 89‑96.［张文煜，马可可，郭振海，等. 基于灰狼算法和极限学习机的风速多步预测［J］. 郑州大学学报（工学版），2024，45（2）：89‑96.］
[4] Tu Qingyu, Miao Shihong, Lin Yujun, et al. Ultra‑short‑term interval forecasting method for regional wind farms based on dynamic R‑vine Copula model[J]. High Voltage Engineering, 2022, 48(2): 456‑466.［涂青宇，苗世洪，林毓军，等. 基于动态R藤Copula模型的区域风电集群超短期功率区间预测方法［J］. 高电压技术，2022，48（2）：456‑466.］
[5] Zhu Zibin, Meng Anbo, Ou Zuhong, et al. Ultra‑short‑term wind power prediction based on deep ensemble learning model using multivariate mode decomposition and multi‑objective optimization[J]. Modern Electric Power, 2024, 41(3): 458‑469.［朱梓彬，孟安波，欧祖宏，等. 基于多元模态分解与多目标算法优化的深度集成学习模型的超短期风电功率预测［J］. 现代电力，2024，41（3）：458‑469.］
[6] Wu Ni. Research on wind power prediction method based on ensemble model[D]. Wuhan: Hubei University of Technology, 2023.［吴霓. 基于集成模型的风电功率预测方法研究［D］. 武汉：湖北工业大学，2023.］
[7] Xiang Yang. Research on short‑term power prediction method for wind farm clusters based on CNN‑LSTM modeling[D]. Beijing: North China Electric Power University, 2024.［向阳. 基于CNN‑LSTM模型的风电场集群短期功率预测方法研究［D］. 北京：华北电力大学，2024.］
[8] Li Chaofeng, Yuan Shengyao, Wang Lingmei, et al. Ultra‑short‑term wind power prediction for wind farms based on informer[J]. Journal of Shanxi University (Natural Science Edition), 2024, 47(6): 1201‑1210.［李超峰，原升耀，王灵梅，等. 基于Informer的风电场超短期功率预测［J］. 山西大学学报（自然科学版），2024，47（6）：1201‑1210.］
[9] Gu Tingting, Huang Yinglu, Wang Yanan, et al. Res‑Net‑UNet short‑term wind power prediction with introduction of multi‑head attention mechanism[J]. Acta Energiae Solaris Sinica, 2026,47(3):474‑480.［顾婷婷，黄亦露，王亚男，等. 基于多头注意力机制的ResNet‑UNet短期风电功率预测［J］. 太阳能学报，2026，47（3）：474‑480.］
[10] Zhou Lina, Liu Xudong. Research on short term wind power prediction method based on CNN‑LSTM[J]. Journal of Heilongjiang Institute of Technology, 2024, 38(6): 44‑50.［周丽娜，刘旭东. 基于CNN‑LSTM的短期风电功率预测方法研究［J］. 黑龙江工程学院学报，2024，38（6）：44‑50.］
[11] Chen Gang, Shan Jingning, Li Danyang, et al. Research on wind power prediction method based on convolutional neural network and genetic algorithm[C]//Proceedings of the 2019 IEEE Innovative Smart Grid Technologies‑Asia (ISGT Asia). Piscataway: IEEE, 2019: 3573‑3578.
[12] Liu Yong, Yang Xihui, Yan Linzi, et al. Wind power prediction based on CEEMD and Granger causality variable selection[J]. China Measurement & Testing Technology, 2023, 49(4): 98‑105.［刘勇，杨熙卉，燕林滋，等. CEEMD双层分解和Granger因果变量选择风电功率预测［J］. 中国测试，2023，49（4）：98‑105.］
[13] Zheng Yingying, Li Xin, Chen Yanxu, et al. Short‑term wind power forecasting method in extreme weather based on stacking multi‑model fusion[J]. High Voltage Engineering, 2024, 50(9): 3871‑3882.［郑颖颖，李鑫，陈延旭，等. 基于Stacking多模型融合的极端天气短期风电功率预测方法［J］. 高电压技术，2024，50（9）：3871‑3882.］
[14] Zhang Haotian, Wen Mi, Li Jinguo, et al. Data driven time attention convolution wind power prediction model[J]. Acta Energiae Solaris Sinica, 2022, 43(10): 167‑176.［张浩田，温蜜，李晋国，等. 数据驱动的时间注意力卷积风电功率预测模型［J］. 太阳能学报，2022，43（10）：167‑176.］
[15] Rao R M, Liu Jie, Verkuil R, et al. MSA transformer[C]//Proceedings of the 38th International Conference on Machine Learning. Singapore: PMLR, 2021: 8844‑8856.
[16] Zhang Gang, Liu Hongchi, Zhang Jiangbin, et al. Wind power prediction based on variational mode decomposition multi‑frequency combinations[J]. Journal of Modern Power Systems and Clean Energy, 2019, 7(2): 281‑288.
[17] Bai Jiandong, Zhu Jiawei, Song Yujiao, et al. A3T‑GCN: attention temporal graph convolutional network for traffic forecasting[J]. ISPRS International Journal of Geo‑Information, 2021, 10(7): 485.
[18] Ye Lin, Li Yilin, Pei Ming, et al. Combined approach for short‑term wind power forecasting under cold weather with small sample[J]. Proceedings of the CSEE, 2023, 43(2): 543‑554.［叶林，李奕霖，裴铭，等. 寒潮天气小样本条件下的短期风电功率组合预测［J］. 中国电机工程学报，2023，43（2）：543‑554.］
[19] Xu Wu, Fan Xinhao, Shen Zhifang, et al. Short‑term wind power prediction using transformer with multi‑scale feature extraction[J]. Acta Energiae Solaris Sinica, 2025, 46(2): 640‑648.［徐武，范鑫豪，沈智方，等. 多尺度特征提取的Transformer短期风电功率预测［J］. 太阳能学报，2025，46（2）：640‑648.］
[20] Liu Tengfei, Chen Liyue, Fang Jiangyi, et al. SCFNet: fusion framework of external spatial features for spatio‑temporal prediction[J]. Computer Science, 2025, 52(4): 110‑118.［刘腾飞，陈李越，房江祎，等. SCF‑Net：一种面向时空预测的外部空间特征融合框架［J］. 计算机科学，2025，52（4）：110‑118.］
[21] Wang Xiaodong, Li Shanshan, Liu Yingming, et al. Ultra‑short‑term wind power prediction based on variable feature weight[J]. Acta Energiae Solaris Sinica, 2023, 44(2): 52‑58.［王晓东，栗杉杉，刘颖明，等. 基于特征变权的超短期风电功率预测［J］. 太阳能学报，2023，44（2）：52‑58.］
[22] Pu Xiaoyun, Yang Jing, Yang Xing, et al. IZOA‑Transformer‑BiGRU short‑term wind power prediction based on decomposition technique[J]. Electronic Measurement Technology, 2025, 48(2): 39‑48.［蒲晓云，杨靖，杨兴，等. 基于分解技术的IZOA‑Transformer‑BiGRU短期风电功率预测［J］. 电子测量技术，2025，48（2）：39‑48.］

更新日期/Last Update: 2026-09-03